Processing M57

[sreilly]

Most of my images have been galaxies but now nebula season is fast approaching and I have acquired some data on M57. The issue I see with this type of bright nebula is keeping the nebula from being blown out while bringing up the background. I know how to do this in PS using layers and so on but am a bit stumped about how to go about doing this in PI. I've combined and color corrected the rgb image and posted it as a downloadable .fit file . This will get you to the file if you are inclined to try. http://www.astral-imaging.com/M57-2rgb_c_bn_cc.fit Any suggestions would be very welcomed.

[Nocturnal]

Hi Steve,

I don't really see how galaxies and nebulae are different in this regard. In both cases we have to compress the high dynamic range of the target to fit in our 8 bit display range. The way to do this is with a non-linear stretch using the histogram process. You spread out the dim parts over a larger range than the bright values. This results in dim areas becoming brighter while the bright areas don't blow out. Don't touch the white point and don't slide the black point too far to the right. You can see in the histogram tool how many pixels you're nulling out. I keep that number below 100 or even below 10.

For some nebulae you can then also use HDRWT to restore some detail in the bright areas that was compressed to a very narrow range.

[sreilly]

Sander,

That's what I tried initially but M57 is so much brighter than the background stars that most are lost  if I try to keep M57 down in brightness. I can use histogram stretch and clearly see the outer two shells and the background galaxy but the central part of the nebula is entirely blown out then. If you have the bandwidth, download the file and you can see what I mean. This file does not have the luminance data added yet. The color is binned 2x2 while the luminance is 1x1.

[Nocturnal]

Hi Steve,

I'll have a look at your image later.

[Ioannis Ioannou]

But if you stretch it as is, with a midtone about 0.06 you are getting the background and the nebula is not really blown out (before you have to backgroundneutralize and colorcalibrate though). The information is still there although in the image it seems to be burned - it is not. Then you have to use HDRWT to compress the dynamic range and/or curves with a good mask, before proceed with eg ACDNR

Just my two cents

[Philip de Louraille]

I just played with the script MaskedStretch which is what I use when I have a case like yours. It still needs a lot of work (wavelet...) but maybe a step in the right direction?